An accurate extraction method of target frequency band signal based on proportional interpolation method

Abstract

A method for accurately extracting target frequency band signals based on proportional interpolation method. First, the target frequency band is filtered to obtain the target frequency band signal components of the original vibration signal; then an amplitude threshold is set for the full frequency spectrum of the original vibration signal, and the amplitude in the spectrum is greater than the threshold. Find the frequency and vibration amplitude corresponding to the maximum point; calculate the average and median of the frequency amplitude corresponding to the maximum point, and filter out the frequencies whose amplitude is greater than or corresponding to the full frequency band. Proportional interpolation spectrum correction is performed on each component, and the components located in the target frequency band and outside the target frequency band are reconstructed; the spectral leakage of the reconstructed frequency components is calculated, the result of filtering the original vibration signal is compensated, and the target frequency band is used. The internal and external spectrum leakage is subjected to vector operation on the target frequency band cut-off filtering of the original vibration signal to obtain the target frequency band signal without the influence of the end effect; the invention can effectively suppress the end effect after filtering.

Description

technical field [0001] The invention belongs to the field of mechanical equipment diagnosis, and in particular relates to a method for accurately extracting a target frequency band signal based on a proportional interpolation method. Background technique [0002] In the field of mechanical equipment fault diagnosis, vibration signals are often used to evaluate equipment operating conditions. Due to differences in equipment structure, faulty components and types, the fault characteristics of different equipment are located in different frequency bands, so the specific frequency bands of vibration signals are often extracted and analyzed according to the actual analysis situation. Filtering is a commonly used method for extracting the target frequency band. However, if the leakage problem of the target frequency band is ignored and the target frequency band is directly filtered, it is easy to cause the end effect of the time domain signal of the filtered signal and the distort...

AI Technical Summary

Problems solved by technology

Due to the complex frequency components in the measured signal, the sampling time of the signal is often difficult to satisfy the period of all signal components at the same time, so there must be frequency components with spectrum leakage in the measured signal

Images